TMA Navigator

TMA Navigator analyzes tissue microarray (TMA) data to estimate biomarker protein expression in tumor biopsies and to support patient stratification, survival analysis, and construction of tissue- and cell-type-specific molecular networks.

Key Features:

• Expression data support: Accepts categorical, semi-continuous, and continuous expression scores for TMA datasets.
• Batch-effect correction: Applies an enhanced ComBat algorithm tailored for automated correction of non-biological variation in TMA data.
• Unsupervised grouping: Performs Gaussian mixture modeling with Bayesian information criterion (BIC) regularization to select the optimal number of sample groups.
• Survival analysis: Implements Kaplan-Meier survival estimation with group comparisons using the Mantel-Cox log-rank test.
• Network inference: Facilitates construction of tissue- and cell-type-specific networks from TMA expression data, including analyses designed for limited marker panels.

Scientific Applications:

• Biomarker quantification: Estimation of biomarker protein expression in tumor biopsies using TMA-derived scores.
• Patient stratification: Identification of patient subgroups based on marker score distributions via mixture modeling.
• Prognostic analysis: Comparison of survival outcomes across groups to evaluate prognostic markers using Kaplan-Meier and Mantel-Cox tests.
• Network biology: Derivation of tissue- and cell-type-specific molecular networks from TMA expression data to investigate mechanisms underlying pathophenotypes.

Methodology:

Methods explicitly include enhanced ComBat for batch correction, Gaussian mixture modeling with BIC regularization for unsupervised grouping, Kaplan-Meier survival analysis with Mantel-Cox log-rank test, and network inference approaches for constructing tissue- and cell-type-specific networks.